Stability has improved tremendously from the earlier version and
it is far more econonical on memory.
Now I just get a few of these show up intermittently while running
on 70 fibre drives on 5 qla2200 controllers.
wait_on_irq, CPU 0:
irq: 1 [ 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 ]
bh: 1 [ 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ]
Call Trace: [<e0000000044974d0>] sp=0xe000000048b7fa60 bsp=0xe000000048b79928
decoded to show_stack [kernel] 0x50
[<e000000004494030>] sp=0xe000000048b7fc20 bsp=0xe000000048b798f8
decoded to show [kernel] 0x310
[<e000000004494560>] sp=0xe000000048b7fc20 bsp=0xe000000048b798a8
decoded to __global_cli [kernel] 0x3e0
[<e00000000473e160>] sp=0xe000000048b7fc20 bsp=0xe000000048b79868
decoded to rs_timer [kernel] 0xa0
[<e000000004502110>] sp=0xe000000048b7fc20 bsp=0xe000000048b797f0
decoded to __run_timers [kernel] 0x290
[<e000000004503100>] sp=0xe000000048b7fc20 bsp=0xe000000048b797c0
decoded to run_all_timers [kernel] 0xc0
[<e0000000044f8f90>] sp=0xe000000048b7fc20 bsp=0xe000000048b797a0
decoded to bh_action [kernel] 0x110
[<e0000000044f8bb0>] sp=0xe000000048b7fc20 bsp=0xe000000048b79750
decoded to tasklet_hi_action [kernel] 0x250
[<e0000000044f80e0>] sp=0xe000000048b7fc20 bsp=0xe000000048b796d8
decoded to do_softirq [kernel] 0x140
[<e000000004496e80>] sp=0xe000000048b7fc20 bsp=0xe000000048b796b0
decoded to ia64_handle_irq [kernel] 0x100
[<e000000004491da0>] sp=0xe000000048b7fc20 bsp=0xe000000048b796b0
decoded to ia64_leave_kernel [kernel] 0x0
[<e0000000048e6cd0>] sp=0xe000000048b7fdc0 bsp=0xe000000048b79510
decoded to long_copy_user [kernel] 0xd0
[<e0000000046a2e00>] sp=0xe000000048b7fdc0 bsp=0xe000000048b79440
decoded to __pagebuf_do_delwri [kernel] 0x220
[<e0000000046a32c0>] sp=0xe000000048b7fdd0 bsp=0xe000000048b79370
decoded to _pagebuf_file_write [kernel] 0x260
[<e0000000046a36c0>] sp=0xe000000048b7fe10 bsp=0xe000000048b79280
decoded to pagebuf_generic_file_write [kernel] 0x2c0
[<e0000000046aee90>] sp=0xe000000048b7fe20 bsp=0xe000000048b791d0
decoded to xfs_write [kernel] 0x390
[<e0000000046a6350>] sp=0xe000000048b7fe20 bsp=0xe000000048b79178
decoded to linvfs_write [kernel] 0x5d0
[<e000000004549db0>] sp=0xe000000048b7fe60 bsp=0xe000000048b79100
decoded to sys_write [kernel] 0x210
[<e000000004491d80>] sp=0xe000000048b7fe60 bsp=0xe000000048b79100
decoded to ia64_ret_from_syscall [kernel] 0x0
Then I just need a patch to eliminate the lru_list_lock...
On Thursday 21 March 2002 14:32, Nathan Straz wrote:
> On Thu, Mar 21, 2002 at 01:57:36PM -0700, Warren Stockton wrote:
> > What compiler would you recommend for ia64?
> We are still building ia64 kernels with 2.96 (-85 or -101) and we
> haven't found any compiler related issues yet.
> David Mosberger is compiling ia64 kernels with gcc 3.0, but I doubt he's
> compiling XFS in his kernels. The general concensus on the linux-ia64
> list is that gcc 3.1 doesn't compile bootable ia64 kernels.
> So all I can recommand now is staying with 2.96. I would like to hear
> any success stories with 3.0 or later compilers.